Novel Prediction of Total Fat Content in Cocoa Beans by FT-NIR Spectroscopy Based on Effective Spectral Selection Multivariate Regression

Total fat content is a major quality parameter that chocolate manufactures consider when selecting cocoa beans. This paper attempted the feasibility of measuring total fat content in cocoa beans by using Fourier transform near-infrared (FT-NIR) spectroscopy based on a novel systematic study on efficient spectral variables selection multivariate regression. After the efficient spectra interval selection by synergy interval partial least squares (Si-PLS), the data were treated with support vector machine regression (SVMR) leading to synergy interval support vector machine regression (Si-SVMR). Experimental results showed that the model based on the novel Si-SVMR algorithm was superior to the others. The optimum results were assessed by root-mean-square error of prediction (RMSEP) and correlation coefficient (R _pre) in the prediction set. The performance of Si-SVMR model was RMSEP?=?0.015 and R _pre?=?0.9708. This study has demonstrated that the total fat content in cocoa beans could rapidly be predicted by FT-NIR spectroscopy and Si-SVMR technique. The novel strength and accuracy of Si-SVMR in contrast to other multivariate algorithms has been derived.     

Feasibility study on the use of Fourier transform near-infrared spectroscopy together with chemometrics to discriminate and quantify adulteration in cocoa beans

Fourier transform near-infrared (FT-NIR) spectroscopy combined with Support Vector Machine (SVM) and synergy interval partial least square (Si-PLS) was attempted in this study for cocoa bean authentication. SVM was used to develop an identification model to discriminate between fermented cocoa beans (FC), unfermented cocoa beans (UFC) and adulterated cocoa bean (5–40 wt/wt.% content of UFC). Si-PLS model was used to quantify the addition of UFC in FC. SVM model accurately discriminated the cocoa bean samples used. After cross-validation, the optimal identification rate was 100% in both the training set and prediction set at three principal components. For quantitative analysis, Si-PLS model was evaluated according to root mean square error of prediction (RMSEP) and coefficient of correlation in prediction (R_pred). The results revealed that Si-PLS model in this work was promising. The optimal performance of Si-PLS model showed an excellent predictive potential, RMSEP = 1.68 and R_pred = 0.98 in the prediction set. The overall results indicated that FT-NIR spectroscopy together with an appropriate multivariate algorithm could be employed for rapid identification of fermented and unfermented cocoa beans as well as the quantification of UFC down to 5% in FC for quality control management.     

Detection of Aspartic Acid in Fermented Cordyceps Powder Using Near Infrared Spectroscopy Based on Variable Selection Algorithms and Multivariate Calibration Methods

Near infrared (NIR) spectroscopy combined with chemometrics was employed to detect the aspartic acid content in fermented Cordyceps powder. One hundred sixty-nine samples were applied for calibration (n?=?113) and prediction (n?=?56). Six different pretreatment methods were compared to determine the optimal pretreatment method for analysis. The wavelength variables selected by regression coefficient analysis, successive projections algorithm, and genetic algorithm–partial least squares analysis (GAPLS) were used as the inputs of partial least-squares analysis, multiple linear regression (MLR), and least-squares support vector machine. The performances of these calibration methods were also compared to determine the best model. The results indicated that GAPLS-MLR obtained the highest precision with a correlation coefficient of prediction r _pre?=?0.9223, root mean square of prediction RMSEP?=?0.0751 g/100 g, and coefficient of variation CV?=?5.15 %. The overall results showed that NIR was feasible for the determination of aspartic acid in fermented Cordyceps powder and GAPLS could perform well with less input dimension and computation complexity in the aspartic acid estimation.     

Application of Near-Infrared Hyperspectral Imaging with Variable Selection Methods to Determine and Visualize Caffeine Content of Coffee Beans

Hyperspectral imaging covering the spectral range of 874–1734 nm was used to determine caffeine content of coffee beans. Spectral data of 958.24–1628.89 nm were extracted and preprocessed. Partial least squares regression (PLSR) model on the preprocessed full spectra obtained good performance with coefficient of determination of prediction (R ² _p ) of 0.843 and root mean square error of prediction (RMSEP) of 131.904 μg/g. In addition, 10 variable selection methods were applied to select the best optimal wavelengths. The PLSR models on the different optimal wavelengths obtained satisfactory results. The PLSR model on the wavelengths selected by random frog (RF) performed the best, with R ² _p of 0.878 and RMSEP of 116.327 μg/g. The RF wavelength selection combined with the PLSR model also achieved satisfactory visualization of caffeine content between different coffee beans. The overall results indicated that optimal wavelength selection was an efficient method for spectral data preprocessing, and hyperspectral imaging was illustrated as a potential technique for real-time online determination for caffeine content of coffee beans.     

Prediction of titratable acidity, malic acid, and citric acid in bayberry fruit by near-infrared spectroscopy

An experiment was conducted to simultaneously measure titratable acidity, malic acid, and citric acid of bayberry fruit in a nondestructive manner using near-infrared (NIR) transmittance spectroscopy and chemometrics. The sampling set included different cultivars that were obtainable from different areas in China. Calibration models using partial least squares (PLS) regression were developed based on GB 12293-90 of China and with high-performance liquid chromatography (HPLC) as reference methods. Different preprocessing methods and different wave bands were applied. The correlation coefficient of calibration (r_c), root-mean-square error of calibration (RMSEC), and root-mean-square error of prediction (RMSEP) of the best model for titratable acidity was 0.8959, 2.24, and 2.89 g/L, respectively, with the range of 10,000-5405 cm⁻¹. R_c, RMSEC, and RMSEP values for malic acid and citric acid were 0.6689, 0.32, 0.47 and 0.8970, 1.51, 2.12 g/L, respectively. The prediction accuracies could not be improved by using first and second derivative pretreatment methods. Due to the short time consumption and low monitoring cost, NIR spectroscopic technique has its potential for the rapid and nondestructive prediction of titratable acidity and citric acid in bayberry fruit in a temperature-controlled room, although the accuracy was not high.     

基于光谱技术与光学仿真的柚果在线检测托盘设计与试验

采用Trace pro软件对柚果及配套传送托盘模型进行光学仿真分析,设计一款适用于柚果在线检测的传送托盘。通过结构优化和材料替换等方法,对仿真结果中辉度/照度值较高的托盘模型,进行实物加工。柚果置于不同类托盘上,自主搭建光谱平台分别采集柚果漫透射光谱进行试验对比。以试验结果最优的托盘配合柚果进行光谱采集,光谱经预处理,连续投影算法提取光谱特征波段,建立柚果可溶性固形物含量的偏最小二乘回归法的预测模型,最佳预测决定系数为0.957;预测均方根误差为0.271°Brix。结果表明,模型的预测能力好、预测精度高,传送托盘设计合理。     

Hyperspectral Imaging Sensing of Changes in Moisture Content and Color of Beef During Microwave Heating Process

Moisture content (MC) and color are two important quality parameters of beef during microwave heating process. This study examined the effects of microwave heating time (0–75 s) on MC, color, and myoglobins of beef samples. The results showed that heating time significantly influenced the MC, color (L*, a*), and percentage of related myoglobins. The suitability of hyperspectral imaging (HSI) (400–1000 nm) was investigated to correlate the mean spectra of beef samples and the color and MC values during microwave treatment. After the use of pre-processing methods and optimum wavelengths selection, the SG-SPA-LS-SVM prediction model for MC (R²_P?=?0.869, RMSEP?=?1.304, and RPD?=?2.724) and the SG-RC-MLR model for a* (R²_P?=?0.890, RMSEP?=?0.735, and RPD?=?2.733) were established. The models were then used to develop the distribution maps of MC and a* values, respectively, showing that both MC and a* at the center of the meat slices were higher than those at the edge, corresponding to the temperature distribution during microwave heating. The results demonstrated the ability of HSI system for monitoring the changes of some quality parameters during microwave heating.     

Spontaneous organic cocoa bean box fermentations in Brazil are characterized by a restricted species diversity of lactic acid bacteria and acetic acid bacteria

Spontaneous organic cocoa bean box fermentations were carried out on two different farms in Brazil. Physical parameters, microbial growth, bacterial species diversity [mainly lactic acid bacteria (LAB) and acetic acid bacteria (AAB)], and metabolite kinetics were monitored, and chocolates were produced from the fermented dry cocoa beans. The main end-products of the catabolism of the pulp substrates (glucose, fructose, and citric acid) by yeasts, LAB, and AAB were ethanol, lactic acid, mannitol, and/or acetic acid. Lactobacillus fermentum and Acetobacter pasteurianus were the predominating bacterial species of the fermentations as revealed through (GTG)₅-PCR fingerprinting of isolates and PCR-DGGE of 16S rRNA gene PCR amplicons of DNA directly extracted from fermentation samples. Fructobacillus pseudoficulneus, Lactobacillus plantarum, and Acetobacter senegalensis were among the prevailing species during the initial phase of the fermentations. Also, three novel LAB species were found. This study emphasized the possible participation of Enterobacteriaceae in the cocoa bean fermentation process. Tatumella ptyseos and Tatumella citrea were the prevailing enterobacterial species in the beginning of the fermentations as revealed by 16S rRNA gene-PCR-DGGE. Finally, it turned out that control over a restricted bacterial species diversity during fermentation through an ideal post-harvest handling of the cocoa beans will allow the production of high-quality cocoa and chocolates produced thereof, independent of the fermentation method or farm.     

Acidic Characteristics of Fermented and Dried Cocoa Beans from Different Countries of Origin

Thirty-nine fermented and dried cocoa bean samples from 13 countries were evaluated for pH, titratable acidity and concentrations of volatile and nonvolatile acids. The correlation coefficient between pH and log₁₀ titratable acidity was -0.94. Cocoa beans from Brazil and Far Eastern countries were highly acidic while those from Central American and South American countries were low in acidity. Samples from West African countries were intermediate with titratable acidity values from 0.12 to 0.15 meq NaOH/g sample and pH values from 5.20 to 5.49. Highly acidic beans were characterized by high concentrations of acetic and lactic acids. The high correlation between acetic acid and both pH (r=0.86) and titratable acidity (r = 0.91) indicated that this acid could be primarily responsible for high acidity in cocoa beans.     

On-farm implementation of a starter culture for improved cocoa bean fermentation and its influence on the flavour of chocolates produced thereof

Cocoa bean fermentations controlled by means of starter cultures were introduced on several farms in two different cocoa-producing regions (West Africa and Southeast Asia). Two starter culture mixtures were tested, namely one composed of Saccharomyces cerevisiae H5S5K23, Lactobacillus fermentum 222, and Acetobacter pasteurianus 386B (three heaps and one box), and another composed of L. fermentum 222 and A. pasteurianus 386B (seven heaps and one box). In all starter culture-added cocoa bean fermentation processes, the inoculated starter culture species were able to outgrow the natural contamination of the cocoa pulp-bean mass and they prevailed during cocoa bean fermentation. The application of both added starter cultures resulted in fermented dry cocoa beans that gave concomitant milk and dark chocolates with a reliable flavour, independent of cocoa-producing region or fermentation method. The addition of the lactic acid bacterium (LAB)/acetic acid bacterium (AAB) starter culture to the fermenting cocoa pulp-bean mass accelerated the cocoa bean fermentation process regarding citric acid conversion and lactic acid production through carbohydrate fermentation. For the production of a standard bulk chocolate, the addition of a yeast/LAB/AAB starter culture was necessary. This enabled an enhanced and consistent ethanol production by yeasts for a successful starter culture-added cocoa bean fermentation process. This study showed possibilities for the use of starter cultures in cocoa bean fermentation processing to achieve a reliably improved fermentation of cocoa pulp-bean mass that can consistently produce high-quality fermented dry cocoa beans and flavourful chocolates produced thereof.     

Nondestructive Hygiene Monitoring on Pork Meat Surface Using Excitation–Emission Matrices with Two-Dimensional Savitzky–Golay Second-Order Differentiation

To develop a rapid and nondestructive hygiene-monitoring system in meat-processing plants, plate count and ATP content on pork meat surfaces were quantitatively determined with particular attention to NAD(P)H fluorescence produced by microorganisms. An excitation (Ex)–emission (Em) matrix (EEM) was obtained, and the five fluorescence peaks of tryptophan, NAD(P)H, zinc protoporphyrin IX, protoporphyrin IX, and riboflavin were observed. Plate count and ATP content were predicted with good accuracy [r _p?=?0.90–0.94 and root mean square error of prediction (RMSEP)?=?log₁₀ (0.68–0.79 CFU cm^?2) for plate count and r _p?=?0.84–0.89 and RMSEP?=?log₁₀ (0.61–0.71, mol cm^?2) for ATP content]. Two-dimensional Savitzky–Golay second-order differentiation was found to be a powerful preprocessing tool of EEMs to improve prediction accuracy. Better prediction accuracy was obtained when the sensitivity of the fluorescence spectrophotometer was set to focus on fluorescence from NAD(P)H than that from both tryptophan and NAD(P)H. However, little linear relationship was observed between plate count and fluorescence intensity from NAD(P)H (R ²?=?0.31). The absolute value of regression coefficient (RC) of partial least-squares regression (PLSR) at the wavelength assigned to NAD(P)H, zinc protoporphyrin IX, protoporphyrin IX, and riboflavin was high. It can be concluded that a good prediction model was developed in which four fluorescence compounds of NAD(P)H, zinc protoporphyrin IX, protoporphyrin IX, and riboflavin contribute to the prediction model, and these compounds are probably produced by microorganisms.     

Discrimination of Cocoa Beans According to Geographical Origin by Electronic Tongue and Multivariate Algorithms

Electronic tongue as an advanced and novel emerging technology has been successfully utilized for the rapid identification of cocoa beans according to their geographical locations. Seven categories of cocoa beans from Ghana were used in this experiment. Electronic tongue system was used for data acquisition while three pattern recognition methods were applied comparatively to build discrimination model. The performances of the models were cross-validated to ensure its stability. Experimental results revealed that Fisher’s discriminant analysis (FDA) is better than principal component analysis (PCA) for visualizing the cluster trends. K-nearest neighbor (KNN) model was slightly better than FDA model at an optimal performance of 100 % in the training set and 98.8 % in prediction set. Overall, support vector machine (SVM) was superior to both FDA and KNN with 100 % discrimination rate in both the training and prediction set at five PCs. This finding proves that electronic tongue technology coupled with SVM technique can rapidly, accurately, and reliably discriminate cocoa beans for quality assurance management.     

Variations of peroxidase activity in cocoa (Theobroma cacao L.) beans during their ripening,fermentation and drying

An increase of peroxidase activity in the seeds of cocoa (Theobroma cacao L.) during their ripening has been determined. An additional increase of the peroxidase activity (about 10 times) is observed during the fermentation and drying of the beans. The residual activity determined in the cocoa beans after sun-drying was higher than that in unfermented ripe seeds. The major cocoa isoperoxidase was shown to be an acidic enzyme with pI 4.7. Using isoelectrofocusing, the appearance of two basic isoenzymes of the peroxidase with pI 8.6 and 9.0 during the process of the fermentation has been detected. ©     

Ghanaian Cocoa Bean Fermentation Characterized by Spectroscopic and Chromatographic Methods and Chemometrics

Abstract: Export of cocoa beans is of great economic importance in Ghana and several other tropical countries. Raw cocoa has an astringent, unpleasant taste, and flavor, and has to be fermented, dried, and roasted to obtain the characteristic cocoa flavor and taste. In an attempt to obtain a deeper understanding of the changes in the cocoa beans during fermentation and investigate the possibility of future development of objective methods for assessing the degree of fermentation, a novel combination of methods including cut test, colorimetry, fluorescence spectroscopy, NIR spectroscopy, and GC-MS evaluated by chemometric methods was used to examine cocoa beans sampled at different durations of fermentation and samples representing fully fermented and dried beans from all cocoa growing regions of Ghana. Using colorimetry it was found that samples moved towards higher a* and b* values as fermentation progressed. Furthermore, the degree of fermentation could, in general, be well described by the spectroscopic methods used. In addition, it was possible to link analysis of volatile compounds with predictions of fermentation time. Fermented and dried cocoa beans from the Volta and the Western regions clustered separately in the score plots based on colorimetric, fluorescence, NIR, and GC-MS indicating regional differences in the composition of Ghanaian cocoa beans. The study demonstrates the potential of colorimetry and spectroscopic methods as valuable tools for determining the fermentation degree of cocoa beans. Using GC-MS it was possible to demonstrate the formation of several important aroma compounds such 2-phenylethyl acetate, propionic acid, and acetoin and the breakdown of others like diacetyl during fermentation. Practical Application: The present study demonstrates the potential of using colorimetry and spectroscopic methods as objective methods for determining cocoa bean quality along the processing chain. Development of objective methods for determining cocoa bean quality will be of great importance for quality insurance within the fields of cocoa processing and raw material control in chocolate producing companies.     

Evaluation of Canning Quality Traits in Black Beans (Phaseolus vulgaris L.) by Visible/Near-Infrared Spectroscopy

Black bean (Phaseolus vulgaris L.) processing presents unique challenges because of discoloration, breakage, development of undesirable textures, and off-flavors during canning and storage. These quality issues strongly affect processing standards and consumer acceptance for beans. In this research, visible and near-infrared (Vis/NIR) reflectance data for the spectral region of 400–2,500 nm were acquired from intact dry beans for predicting five canning quality traits, i.e., hydration coefficient (HC), visual appearance (APP) and color (COL), washed drained coefficient (WDC), and texture (TXT), using partial least squares regression (PLSR). A total of 471 bean samples harvested and canned in 2010, 2011, and 2012 were used for analysis. PLSR models based on the Vis/NIR data showed low predictive performance, as measured by correlation coefficient for prediction (R _pred) for APP (R _pred?=?0.275–0.566) and TXT (R _pred?=?0.270–0.681), but better results for predicting HC (R _pred?=?0.517–0.810), WDC (R _pred?=?0.420–0.796), and COL (R _pred?<?0.533–0.758). In comparison, color measurements from a colorimeter on drained canned beans showed consistently good predictions for COL (R _pred?=?0.796–0.907). In spite of the low or relatively poor agreement among the sensory panelists as determined by multirater Kappa analysis (K _free of 0.20 for APP and 0.18 for COL), a linear discriminant model using the Vis/NIR data was able to classify the canned bean samples into two sensory quality categories of “acceptable” and “unacceptable”, based on panelists’ ratings for APP and COL traits of canning beans, with classification accuracies of 72.6 % or higher. While Vis/NIR technique has the potential for assessing bean canning quality from intact dry beans, improvements in sensing and instrumentation are needed in order to meet the application requirements.     

Hyperspectral imaging as an effective tool for prediction the moisture content and textural characteristics of roasted pistachio kernels

The objective of this study was to develop calibration models for prediction of moisture content and textural characteristics (fracture force, hardness, apparent modulus of elasticity and compressive energy) of pistachio kernels roasted in different conditions (temperatures 90, 120 and 150 °C; times 20, 35 and 50 min and air velocities 0.5, 1.5 and 2.5 m/s) using Vis/NIR hyperspectral imaging and multivariate analysis. The effects of different pre-processing methods and spectral treatments such as normalization [multiplicative scatter correction (MSC), standard normal variate transformation (SNV)], smoothing (median filter, Savitzky–Golay and Wavelet) and differentiation (first derivative, D¹ and second derivative, D²) on the obtained data were investigated. The prediction models were developed by partial least square regression (PLSR) and artificial neural network (ANN). The results indicated that ANN models have higher potential to predict moisture content and textural characteristics of roasted pistachio kernels comparing to PLSR models. High correlation was observed between reflectance data and fracture force (R²?=?0.957 and RMSEP?=?3.386) using MSC, Savitzky–Golay and D¹, compressive energy (R²?=?0.907 and RMSEP?=?15.757) using the combination of MSC, Wavelet and D¹, moisture content (R²?=?0.907 and RMSEP?=?0.179) and apparent modulus of elasticity (R²?=?0.921 and RMSEP?=?2.366) employing combination of SNV, Wavelet and D¹, respectively. Moreover, Vis–NIR data correlated well with hardness (R²?=?0.876 and RMSEP?=?5.216) using SNV, Wavelet and D². These results showed the capability of Vis/NIR hyperspectral imaging and the central role of multivariate analysis in developing accurate models for prediction of moisture content and textural properties of roasted pistachio kernels.     

Mass transfer of water and volatile fatty acids in cocoa beans during drying

In order to elucidate the phenomena involved in the remnant acidity of cocoa beans dried artificially, the diffusivities of water, and volatile fatty acids (VFAs: acetic, propionic, butyric and iso-butyric acids) in cocoa beans during drying were evaluated. Experimental drying kinetics of the acids were conduced at 40–60 °C with and without shell. Samples were taken at different drying times for moisture and acids content evaluation. VFAs content was evaluated by GC in a methanolic extract, and moisture content by a vacuum oven. Mass diffusivity was estimated from the fitting of experimental kinetics to a theoretical model that takes into consideration the beans’ shape. Acetic, propionic and butyric acids diffusivities were significantly (p < 0.05) smaller than water diffusivities both with and without shell. VFAs diffusivities were between 1/6 and 1/22 diffusivities values for water. Iso-butyric acid diffusivity was not statistically significant but the value was smaller than for the other VFAs. The diffusivities of VFAs may explain the remnant acidity in artificially dried cocoa beans.     

Detection of 5-HMF in apple juice with artificial sensing systems

5-HMF (5-hydroxymethyl-furfural) is a product of thermal treatment and is increasingly considered a food contaminant. Here, different concentrations of 5-HMF were measured in apple juice to evaluate the performance of the electronic nose (EN) and electronic tongue (ET) as rapid detection techniques for 5-HMF when coupled with chemometric analysis. Principal component analysis (PCA) and linear discriminant analysis (LDA) evaluated the discrimination capacity of EN and ET for 5-HMF. Loading analysis examined the discrimination contribution of the EN sensors. Partial least square (PLS) regression analysis established a quantitative prediction model for different concentrations of 5-HMF based on EN and ET data set. The optimal models had a coefficient of determination (R²) of 0.926 and a root-mean-square error of prediction (RMSEP) of 0.4168 in EN; there were R² of 0.914 and RMSEP of 0.5836 in ET. These results demonstrate that EN and ET coupled with chemometric analysis are two promising approaches for the rapid and online detection of 5-HMF in apple juice.     

Development of Color Indicators for Kimchi Packaging

Color indicators for kimchi(fermented vegetable product) packaging were developed to monitor the quality of such commercial products during storage and distribution. Kimchiwas packed in a polypropylene (PP) tray with nylon/cast polypropylene (Ny/CPP) lid. A color indicator film consisting of PP resin, Ca(OH)₂ as a CO₂ absorbent and bromocresol purple or methyl red was attached to the lid. During fermentation at 0,10, and 20°C, kimchi quality was evaluated for pH and titratable acidity (TA), which had been correlated with Hunter color values of the indicators. Regardless of fermentation temperature, the color values correlated well with pH and TA values.     

Performance of different fermentation methods and the effect of their duration on the quality of raw cocoa beans

This study aims to compare the effect of three cocoa fermentation methods and their duration on raw cocoa quality. Results showed a decrease in percentage of physical quality defects on fermentation method. Cocoa fermented for 4 days presented higher percentage of purple beans reached 45% and about 10% of slaty beans than cocoa fermented for 6 days whatever the process. Fermentation duration did not influence the mouldy beans that were around 1%. Formation of brown beans increased from 16% to 50% depending on the fermentation duration and process. Using wooden boxes allowed higher percentage of 77%–90% brown beans than others materials. Acidity of cocoa decreased on fermentation duration but beans treated in boxes were significantly (P = 0.05) acidic from 1.40 and 3.07 meq of NaOH g^?1. Fungal population did not vary in number depending both on the duration and the fermentation method with rates that ranged from 3.32 × 10⁷ to 8.63 × 10⁷ CFU g^?1.